The use of generative artificial intelligence technologies in education has the potential to revolutionize learning and educational assessment by personalizing the learning experience, providing immediate feedback, and improving the overall learning experience.The relevance of the research is due to the spread of artificial intelligence technologies and the lack of practices for formative assessment of educational achievements in higher education.The problem statement: existing domestic e-learning systems have limited functionality, which makes them difficult to use in the process of formative testing.The goal of the research is to study the possibility and effectiveness of using generative artificial intelligence technologies in formative assessment in higher education.The objectives of the research are to consider the theoretical foundations of using generative artificial intelligence tools in assessing knowledge, skills and abilities, and to analyze our own experience of using large language models in formative testing at a university.The methodological basis of the research is analysis of Internet resources and literary sources, methods of mathematical statistics, and synthesis.The research results: the possibilities of using and the effectiveness of generative artificial intelligence technologies in formative testing of university students have been studied, our own experience of using large language models in formative testing have been analyzed, the main limitations of introducing these technologies into the educational process have been identified, recommendations have been given for organizing formative testing using large language models.Key conclusions: Large language models can be integrated into the learning process to assess formative and summative tests, which will significantly reduce the workload of teachers, provide more objective results and, ultimately, increase the effectiveness of the learning process.